The latest insights into axon regeneration research
In this article, we throw light on the underlying mechanisms of axon regeneration. Axons play a central role in the recovery of neuronal functions after injury or trauma.
Find out how researchers assess axonal growth with the help of specialized axon quantification software. We reached out to Dr. David Hercher from the Ludwig Boltzmann Institute for Traumatology, The Research Center in Cooperation with AUVA (LBI) and asked him to give us insights on how he and his team cooperated with us in the development of the IKOSA Axon Quantification Application.
Table of contents
Meet Dr. David Hercher and his team
David Hercher studied Molecular Biology at the University of Vienna and finished his Doctorate in Medical Sciences at the Medical University of Vienna. His main research interest is the investigation of nerve regeneration after injuries, with a special focus on the peripheral nervous system and improving outcomes after peripheral nerve injuries with segmental tissue loss. Adding up to his expertise in nerve regeneration research, David has a keen interest in novel imaging modalities and their use in the field of regenerative medicine.
Dr. Hercher has a long list of international publications covering topics like:
Exploring axon regeneration mechanisms on the foreground of neuroscience
Thank you for having me. First of all, I am a molecular biologist by training. I studied at the University of Vienna, where I did my Masters degree. Afterwards I did my Doctorate in Medical Science at the Medical University of Vienna, all while I was working at the Ludwig Boltzmann Institute in Vienna. I took over more and more responsibilities there and am now the head of the Neuroregeneration research group. I supervise a small group of 6-11 people and we investigate anything regarding nerve regeneration or neuronal regeneration.
This is definitely my main research interest, so basically anything that has to do with injuries or processes after injury and that relates to nervous tissue.
I would like to give you a short background on the Ludwig Boltzmann Institute for Traumatology, The Research Center in Cooperation with AUVA (LBI). It was founded in the 1980s by the AUVA (meaning Allgemeine Unfallversicherungsanstalt or the Austrian Workers’ Compensation Board).
Our mission is to develop and advance diagnostic, therapeutic and regenerative research strategies and to provide the best care possible to the patient. This can be achieved by research strategies, where we as researchers come in. We focus on everything that happens after injury: from acute care, septic shock up to long term regeneration and rehabilitation.
Good question. During my molecular biology studies I specialized quite early in neuroscience, immunology and molecular medicine, as they are linked to each other. Then I did the first courses at the Center for Brain Research in Vienna and realized that the brain is a super huge ‘black hole’. At that time, I became really interested in the Peripheral Nervous System (PNS), as I thought back then it was a simpler system, although, the deeper I got into the topic the more I realized that there is nothing simple in biology.
I am researching injuries to the nervous system, because there is a huge clinical need for both the Central Nervous System (CNS) and PNS regeneration. Although the need for novel treatment models of patients with injuries to the nervous system is pressing, the progress the field has made over the last decade is not overwhelming. People sometimes tend to take it for granted to not regain full function or full sensory or motor function after injury. Although there would be possibilities, if we get the funding and do the research.
The PNS and CNS react very differently to injuries or to damage in general. In principle, you have the neurons, which have axons to convey the information and then you have different kinds of glial cells, which provide support and protection to the neurons, maintain homeostasis, clean up debris, and form myelin. Neurons refer to these specialized cells of the nervous system, receiving and transmitting signals, while glial cells are those cells that surround neurons, support and insulate them.
"Myelination or the forming of myelin is important to help provide fast conduction of action potentials. Without myelination the conduction of an impulse from the neuron to, for example, the muscle is very slow", Dr. Hercher explains.
This image shows the four different types of glial cells found in the central nervous system: Ependymal cells (light pink), Astrocytes (green), Microglial cells (red), and Oligodendrocytes (functionally similar to Schwann cells in the PNS) (light blue). Image taken from commons.wikimedia.org.
However, some of the pain fibers are non-myelinated. Here an example: If you step onto a lego barefooted, you can feel this sharp initial stinging pain. That would be the myelinated fibers conveying information to the brain that you just stepped onto a very sharp object and also triggering the removal of the foot off the object. Afterwards there is this throbbing deeper pain, which comes delayed, these would be the unmyelinated fibers.
Anyhow, so you have the glial cells, which myelinate, and there you have the biggest difference between CNS and PNS. In the PNS you have the Schwann cells, which are very plastic cells, whereas in the CNS you have the Oligodendrocytes, which are not as plastic and also react differently to injury.
In general, Oligodendrocytes do not provide a great environment for axons to regenerate once you have a cut or injury. They do not develop into a pro-repair-phenotype, while the Schwann cells in the PNS do exactly that. They totally switch their phenotype and they stop producing myelin in an instant. They even start to eat up the myelin, because myelin is hindering axonal regeneration. Then they proliferate, migrate and form highways for the axons to regrow and give the axons the cues they need in order to find their way, while the CNS very often reacts to injuries by formation of scar tissue.
This scar formation is not unimportant, because you have bleeding after injury and the tissue tries to limit the damage, but the scar is not resolved and then there is a huge obstruction for regrowing axons in, for example, the spinal cord.
You also have chronic inflammation in the spinal cord, whereas in the PNS this inflammation usually is transient. You have pro-inflammation, which you need for regeneration. For example, if you have scratched your leg or cut your leg, you get reddish skin and a crust, which becomes itchy. These are necessary processes of inflammation to heal. If the damage is not too severe, this works quite well.
Actually, it is a very good question. For example, you cut yourself with a knife in the kitchen and the nerve in your finger gets damaged. If you cut the nerve, you have regeneration and all these processes going on at the side of injury. But you have to remember that the sensory neuron, which is responsible for feeling in the fingertip, is not anywhere near there.
The sensory neuron is next to the spinal cord. The axon that it has, reaches from the spinal cord all the way down the arm to your fingertip. The neuron or cell body is also reacting to this injury. They notice the injury and they stop their normal function of conveying and forwarding information in the case of sensory neurons and start to switch on a repair-protocol.
If the brain does not get information from a certain area anymore, then it can’t handle the situation very well. This can be one issue for patients who experience neuropathic pain or phantom pain. In that case a limb, which has been amputated, is hurting the patient. It is still not completely understood how this works.
My main interest lies in what Schwann Cells do. They are very fascinating cells, in my opinion, as they switch from one phenotype to a totally different phenotype and have a complex interplay with other cells such as endothelial cells for blood vessels, lymphatic cells, macrophages (inflammatory cells in the PNS).
There are many factors involved in this process of axon regeneration. If you have a segmental injury with two separated nerve stumps, it is necessary for the body to form a fibrin bridge between these in order to allow axon growth from the proximal to the distal end. Only afterwards, Schwann cells and endothelial cells grow in. To me this is very fascinating because there are different factors and cells, which interact with each other in order to facilitate nerve repair and in the end, hopefully, reinnervation of the target organ (skin, muscle, inner organ etc.). I would say for me, Schwann cells are fascinating, and if I have to single out one topic, it would be them.
If we define cells as a factor, then for us it would be the lymphatic system, which has gained a lot of interest recently. So, there is nothing out there yet describing the effects of the lymphatic system and its role in potential nerve regeneration.
"We observed very recently that after injury, we have an increase of lymphatics in the peripheral nerve", Dr. Hercher emphasizes.
Usually, you do not have any lymphatics in the inside of a peripheral nerve, but after injury this system has to come in because drainage and immune surveillance are needed, especially when you do not have a clean cut.
However, this is all pre-clinical and we do not know if this holds true to humans as well. But it is very interesting for us, because two new systems come together: the Schwann cells and how they react to lymphatics and vice versa. This would be a very recent factor, which might play a role, but we do not know yet.
Subscribe to our newsletter!
On experimental design, image acquisition and quantification analysis methods for evaluating axonal and neuronal growth
It is quite an old technique called histomorphometry.
You harvest the tissue of interest, in this case the peripheral nerve or the regenerating nerve. Then you stain the myelin sheath, which only gets stained if it is a regenerated axon. This signifies a successful regeneration. Then you can also investigate the thickness of the myelin sheath, how well it is myelinated, how many axons you have there and so on.
However, for me, this also holds a drawback because quite a large portion of cells or axons are unmyelinated or very thinly myelinated. With this technique you miss all of those, which are referred to as c-fibers or autonomic fibers.
For this reason, we chose to go a different route, because we wanted to stain them for neurofilament, which is an intermediate filament inside of the axon. And then we don't miss any of the axons. However, people use this often with fluorescence microscopy, which means you get only one signal, and then you have a neurofilament positive structure, which is cool, because you can also have tools out there to automatically quantify these as well.
But then you also have a loss of information because as everything else is black, you don't get any information on the structural integrity, and you also can't do any other histological analysis with it. You missed this information due to the fluorescence based method. Therefore, we did normal light microscopy and used HRP (Horseradish peroxidase) conjugated antibodies in order to visualize our neurofilament, which gives you this brownish color. With this you get information about the whole environment. However, it is not as easy to automate quantification on these kinds of histological slices.
Once we have our slices, we scan them. Previously we had to manually sit at the microscope and look at them individually to note things down. This was a very time-consuming process, but luckily now there are newer options out there, where you can actually automatically scan your slices to have them digitized.
"Digitalization is the key, obviously, especially if you want to do quantification and not only qualitative assessment", Dr. Hercher states.
And this is, I guess, where also KML Vision and the IKOSA Prisma Axon Quantification App came into play.
On the benefits of the IKOSA Prisma Axon Quantification App
As I said, we wanted to actually automate this process and we tried to use ImageJ deep learning plugins, but nothing turned out the way we wanted to. So, we actually reached out to your company and we got introduced to the IKOSA Platform.
"And together with your computer vision team, we developed this Axon Quantification App, which allows us now, with a very high reproducibility and reliability, to automatically quantify these axons in the 1000s", Dr. Hercher says.
I would say the biggest advantage we took out of it was that we could do it remotely. The platform works very well too, and you can share it with other co-researchers.
But most importantly, it saves a lot of time. So you get quantitative information. Prior to that, you would just put a couple of pictures into your publication. But that's not hard quantitative data. We wanted to have robust, positive data. And this is how we achieved it.
Basically, I did my fair share of manually counting axons and I did not want to put my students through that process, and I also was not happy with the results of automatic quantification from other systems.
"Big projects were coming up and I wanted to have reliable software for axon quantification", Dr. Hercher says.
I need to stress the fact that axon quantification is not alone the marker for regeneration, because a lot of axons does not mean that your regeneration works well, because neurons also sprout. One axon sprouts into several axons after injury. You have to be careful when interpreting these results/readouts. People tend to forget that. But if you do not have any axons, obviously there is no regeneration.
For example, we get parameters such as the area, the circumference and elongation. These are important because axons do not always grow like highways, they grow tilted and take turns.
Another important point is that you do not always cut the nerve straight. When this happens you have a seemingly larger area, because you did not cut it in a 90-degree angle. It is important for us to consider this as we want to be sure we get information of the total area to draw conclusions of the total area to the original fiber (where does it come from and where does it want to go).
With all the other parameters that the Axon Quantification App gives you, you can rule out some false positive or false negative readouts. Generally, it is really straightforward, as you download an excel sheet and you get all these parameters, either for a ROI or the whole picture.
For us the main readout-parameter would be the axon count, because this gives us the idea, if there is axon regeneration or not. As I said, one has to be careful in the interpretation of this data. However, it still is the main outcome parameter. There are additional parameters like the area, which gives you information on the fibers. Larger diameter axons have a higher conduction velocity, which means they are able to send signals faster than small diameter axons. If we know that the axons are actually regenerated ones, we can draw conclusions regarding their function. We can order them into fast or slow conducting fibers.
However, we are not yet capable of differentiating between sprouting and actual regeneration. To do so, we would need further staining. For example, additional staining for myelin or also a lot of cross-sections trying to follow up on one single axon fiber, which is a tough thing to do. You would need maybe even electron-microscopy.
At this point functional evaluation i.e. nerve conduction comes into place with muscular or sensory feedback. You want the axons to reach their respective target.
There is this huge process and it is still not clear how axons find their way in a segmental damage. In humans, regrowth is approximately 1mm/d. If you injure your brachial plexus and you want to reinnervate your fingertips, this would take up to 1000 days to regenerate. So the goal is to also speed up the process (improve regenerative rate) and find ways how we could maybe trigger the pathfinding.
Discover the Axon Quantification App in IKOSA
General app description
the detection and quantification of axons in histological peripheral nerve images
axon count, axon area, axon perimeter, elongation, circularity, mean diameter
I would say definitely we are on the way to get more robust data on the type of fibers, which are regenerating (motor, sensor, autonomic). I also think there is huge progress in 3D visualization which will hopefully allow in vivo tracking of regeneration. There is also some data out there on ultrasound and high resolution ultrasound.
It would be amazing, if we could actually watch the nerve regenerate. So I think digitalization and 3D imaging would definitely be the ones in the axon quantification.
Regarding my research, we do basic research and we want to know which factors are upregulated, when and where and for how long basically. But we also strive for novel treatment models for regenerative processes in general or drugs which would help us improve regeneration.
Another important factor to consider is the needs of the patients and what they wish for. Surveys show that patients are actually not that interested in how heavy they can lift (strength) but rather for how long they can hold/lift it (endurance). Imagine you have a nerve injury, and you have a small child. A small child is not heavy, so you might be able to hold it but maybe only for 10 seconds. It is vital for them that the muscle can be constantly active and does not fatigue. This is super important to consider, what patients need and to also have this driven research.
A success story in nerve regeneration research with the help of IKOSA
We thank Dr.David Hercher and his team for our fruitful cooperation. Two recent articles with the participation of the Ludwig Boltzmann Institute for Traumatology demonstrate the high potential of the IKOSA Platform in neural regeneration research.
Their paper in the MDPI journal Biomolecules showcases how the research team developed a specialized algorithm for the study of lymphatic vessel growth in peripheral nerve autografts with the help of IKOSA.
In an article in the MDPI Biomedicines journal the researchers showcase the automated quantification of axons, lymphatic- and blood vessels following shockwave therapy for the regeneration of the rat median nerve. Again, the data obtained from this novel in-vivo model has been analyzed with a tailor-made IKOSA algorithm.
Seeing our users conduct outstanding research and publish in highly-rated scientific journals motivates us to design bioimage analysis applications for optimal performance.
The next success story could be yours!
Heinzel, J. C., Oberhauser, V., Keibl, C., Schädl, B., Swiadek, N. V., Längle, G., ... & Hercher, D. (2022). ESWT Diminishes Axonal Regeneration following Repair of the Rat Median Nerve with Muscle-In-Vein Conduits but Not after Autologous Nerve Grafting. Biomedicines, 10(8), 1777.
Hromada, C., Hartmann, J., Oesterreicher, J., Stoiber, A., Daerr, A., Schädl, B., ... & Hercher, D. (2022). Occurrence of Lymphangiogenesis in Peripheral Nerve Autografts Contrasts Schwann Cell-induced Apoptosis of Lymphatic Endothelial Cells in vitro.